Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Any news about RC4? (Read 19862 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Any news about RC4?

Just wondering about the target date of RC4 release, is there still much to do?

Any news about RC4?

Reply #2 –
Monty hinted in IRC the other day that a week or so is not out of the question.  So it could be quite soon indeed. In fact pre RC4 code should be in CVS at the moment that can handle 44.1 and 48Khz sterio If I read the chat logs correctly.  Anyone up to compiling?

Any news about RC4?

Reply #3 –
Quote
Originally posted by Neo Neko
In fact pre RC4 code should be in CVS at the moment that can handle 44.1 and 48Khz sterio If I read the chat logs correctly.  Anyone up to compiling?


You mean, right now?

There are some compiles on Rarewares dated 06-24, others dated 06-27. Wouldn't they support 48.000Hz already, maybe?

[span style='font-size:9']Don't ask me for Vorbis tests... [/span]

Any news about RC4?

Reply #4 –
New library changes overnight.  They would appear to be the most significant changes since May.

All new binaries will be posted at Roberto's site later today.

I can't comment on the changes 'cos I haven't even compiled them yet!

Any news about RC4?

Reply #5 –
First impressions - very quick test only!!

Quality as good, or better - approx. 10% reduction in file size!!

OK, all new binaries uploaded to Roberto. Should be available later once he wakes up!!

.exes are smaller; execution time is faster; encoded bitrates/file sizes are smaller; quality (as above) seems similar, or better!

Any news about RC4?

Reply #6 –
Oh my Ogg!! :crystal:
"Taking a jazz approach and concentrating on live playing, I wanted to use several different rhythm sections and vintage instruments and amps to create a timeless sound that's geared more around musicality and vibe than sonic perfection. The key was to write with specific rhythm sections in mind, yet leave open spaces for soloing." Lee Ritenour

Any news about RC4?

Reply #7 –
This is all fine and great =)  I have been bugging iriver some more and here is the response I have received:

"At this time, I can not elaborate as there is no definite release planned.  It is only under advisement with the engineers."

which makes absolutely no sense to me.

Oh well.

-Jeffrey

Any news about RC4?

Reply #8 –
Files uploaded. <yawn>

Regards;

Roberto.

Any news about RC4?

Reply #9 –
I just tested it with -q0, and I have only one word to say about: WOW! :flipoff:

You just have to hear this to believe...

Any news about RC4?

Reply #10 –
Yes, it's amazing! -q0 is very good!  But why aren't the resample and downmix switches working? Are they disabled in this compile? Oggenc does nothing when I use these switches.

Any news about RC4?

Reply #11 –
This is not RC4 yet, but a 'pre-RC4'. There is still code left to check in.
At the moment, only 44.4/48kHz sample rates are working - this will be fixed for the 'real' release.

If you like -q 0, then you should try -q -1, which may or may not make it to the final release  (it's an ~48kb/s mode).

Note that there are still some problems with this code. In particular, 'cse' from OPN #project_mayhem has found a sample (from 'Ventolin' by Aphex Twin) which is obviously non-transparent at -q 5 in either RC3 or the current CVS, but sounds much worse with the current CVS.  You can see what I mean by getting the files from

http://www.maths.warwick.ac.uk/~jingram/listening/

However, overall this release seems like a big step up in quality, which is good news all round.

Any news about RC4?

Reply #12 –
Quote
Originally posted by Jon Ingram
If you like -q 0, then you should try -q -1, which may or may not make it to the final release  (it's an ~48kb/s mode).


It didn't work, it's identical to -q 0 ???

Tested some intense metal tracks at -q0, I'm VERY, VERY impressed!

Any news about RC4?

Reply #13 –
Quote
Originally posted by Jon Ingram
Note that there are still some problems with this code. In particular, 'cse' from OPN #project_mayhem has found a sample (from 'Ventolin' by Aphex Twin) which is obviously non-transparent at -q 5 in either RC3 or the current CVS, but sounds much worse with the current CVS.


Hehe, the sample was found by Dibrom and it was originally for troubleshooting MPC problems, I just played a little with it and tried every encoder I had in hand. Just for the record, this sample is problematic for all encoders, PsyTEL AACenc 2.15 -extreme, LAME --alt-preset standard, MPC --standard and Ogg -q 5 all fail.

Any news about RC4?

Reply #14 –
a big, WOW from my side too. Binaries are now available on my page too!

Any news about RC4?

Reply #15 –
There certainly is a problem of some sort with the resampling. It will permit upsampling, but not downsampling!

Also, more changes in the libs in the CVS today yielding even lower bitrates/file sizes. I'll wait until tomorrow before I do a bulk recompile and upload.

I think the word should be that if you are planning a large encoding session, I should wait a little as after a period of calm there now seems to be a storm of activity!

Any news about RC4?

Reply #16 –
Monty did yesterday the codebook optimizations for the new quality enhancing techniques (noise normalization, new masking curves).

So quality wise the CVS should be pretty much like the rc4 release. Then there's of course bug fixes etc after this. But noise normalization and new masking curves should be fully functioning.

I'm very eager to test low bitrate between Vorbis and MP3Pro VBR. Unfortunately I'm busy until Monday. FF123, can you do some testings?
Juha Laaksonheimo

Any news about RC4?

Reply #17 –
Quote
Originally posted by JohnV
Monty did yesterday the codebook optimizations for the new quality enhancing techniques (noise normalization, new masking curves).

So quality wise the CVS should be pretty much like the rc4 release. Then there's of course bug fixes etc after this. But noise normalization and new masking curves should be fully functioning.

I'm very eager to test low bitrate between Vorbis and MP3Pro VBR. Unfortunately I'm busy until Monday. FF123, can you do some testings?


I probably can't get to it soon either (I'm trying to be good and working on ABX, after wasting too much time on Usenet).  Surely a group test could be appropriate now though, wouldn't you say?  Low bitrates shouldn't really require an ABX mode.

I think wma8 should be included in such a test, though, even though it is unpopular, because I think it is a contender at low bitrates if the basis of comparison is only sound quality.

ff123

Any news about RC4?

Reply #18 –
Vorbis 1.0 should follow pretty soon after rc4 (yeah really ), so I think the group test should be made using 1.0.
Juha Laaksonheimo

Any news about RC4?

Reply #20 –
The old bass bug seems to be back... Only a bit softer this time. bassrumbledemo.wav has the same old knocking up to somewhere between q4 and q5 (170 - 180 kbit/s)... Any ideas why?

Any news about RC4?

Reply #21 –
I haven't messed with Ogg in a while but tried the new pre-RC4 compile, and am pretty impressed. I'm happy with MPC for home use, so I have been playing with Ogg as a possibility for portable use, if the hardware support ever somes along. -q0, while not audiophile quality by any means, is suitable enough for portable use, and doesn't seem to contain the very salient artifacts that WMA does at the same bitrate (~64kbps). I encoded Rush's "Yyz" at -q0 and ended up with a nominal bitrate of 64kbps (average 66). Flanging on the opening triangle taps is not as bad as I remember, Geddy's bass line in the right-channel doesn't warble as bad as it used to, and overall I was surprised that I could actually listen to something at 64kbps without wanting to barf. I would probably use -q2/3 if I had the choice, but in any case, Ogg seems to currently be the best choice in the 64kbps range. Now if only we could get hardware support for Ogg after v1.0, I would be happy

Any news about RC4?

Reply #22 –
Quote
Monty did yesterday the codebook optimizations for the new quality enhancing techniques (noise normalization, new masking curves).


what did he readjust the noise normalization level too? I was reading earlier that he had been normalizing the time domain envelope from an even set to +/- 6 dB of course that could have changed over a long period time, if that's the correct thing I am thinking about.  When you say codebook optimizations do you speak of just regular VQ lattice / Huffman codebooks or the cascaded codebooks? I think he was trying to finish those by the 1.0 release correct?
budding I.T professional

Any news about RC4?

Reply #23 –
Already found my answer to the masking curves question. Still looking around for the noise normalization. I'll fix that.
budding I.T professional

Any news about RC4?

Reply #24 –
The revolution is under way! We are on the road to true free kick arse open multimedia!!!!!!!!!! You have to give the guys at Xiph props! I almost wish it was August already so I could play with 1.0 Vorbis and some Theora alphas! All in good time I suppose.